SICK China's goal is to become a participant in flexible automation, of which robots are a part.
I. Flexible Automation of Robots
Flexible automated production technology, also known as flexible manufacturing technology, is an advanced production technology that automates the processing, manufacturing, assembly, and testing of multiple varieties and batches of products within an enterprise. It is guided by process design and centered on numerical control technology. It involves technologies such as computers, networks, control, information, monitoring, production system simulation, quality control, and production management.
● So what is robot-based flexible automation?
Simply put, it means establishing robotic units so that changes can be made more easily according to needs during the production process.
● Flexible automation has the following requirements:
√ Standardized design
√ Easy to configure
√ May be suitable for products with different characteristics
√ Reliable
● For end users, flexible automation can:
√ Reduce operating costs
√ Reduce investment costs
√ Improve the working environment
● The basic principles of flexible automation in robots:
II. Two robot-based solutions for component grasping:
PLB Vision System:
The PLB is primarily used for locating components within a housing. It includes a 3D camera for acquiring high-quality image data, CAD files of the components for 3D graphic matching, and tools for robot integration. Its function is to enable the robot to pick up objects within the housing.
The image above shows the components of the internal parts picking system. The PLB system includes a 3D vision system for acquiring 3D point cloud data of the box-like structure, and corresponding software. Its main functions include: reading the 3D CAD model data of the gripper and parts, performing collision and overlap calculations; communicating with the robot; configuring the system; selecting the parts to be picked up based on the 3D point cloud data, and outputting the corresponding coordinates to the robot controller.
The following diagram illustrates the scanning range of the PLB vision sensor:
The 3D camera is a ScanningRuler that emits a red laser line to measure the outline of an object. Because it has an internal rotating mechanism that can rotate the laser line at a certain angle, as shown in the figure below, the 3D camera is fixedly installed and can measure the 3D scene below the camera by rotating the internal device.
◆ Minimum distance from the top surface of the object (stand off): 1000 mm
◆ Maximum distance: 2000 mm
◆ Maximum height range: 1000 mm
● (height range + stand - off = max distance)
◆ View of the Field (Volume of View - VOV (LxWxH)):
● 800x1200x1000mm (stand-off is 1 m)
● 1000x1200x750 mm (stand-off measurement: 1.25m)
◆ Scanning speed:
● 3 seconds/image, meaning it takes 3 seconds to scan the entire 3D scene.
◆ Resolution
● 1-4 mm, depending on the distance to the camera
● 1 mm (x,y,z) @ distance to camera is 1m
● 4 mm (x,y,z) @ distance to camera is 2m
◆ Accuracy
● Corresponding to the resolution above
III. PLR Vision System
PLR is a standalone vision system that includes the corresponding software and tools to locate specific parts of components during de-racking, enabling the robot to grasp the parts.
It is based on the principle of 2D vision + 3D vision for positioning. 2D vision is to match images of specific parts, while 3D vision is to emit cross-shaped laser lines to calculate the tilt angle of the part.
The following diagram illustrates the application of a PLR vision system:
◆ Pattern matching (2D) and laser triangulation (3D) principles can calculate the position information (X, Y, Z, Rx, Ry, Rz) of components.
◆ Working distance: 365 mm
◆ Field of view size: 225x170 mm
◆ Positioning accuracy:
● +/- 0.5mm (X,Y,Z)
● +/- 0,1º (Rx, Ry, Rz)
◆ Location time: < 500 ms
The following diagram shows the basic structure of a PLR system:
First, a specific area of the component is learned using a PLR device. During localization, initial localization is performed based on the scaling of the feature areas and the X, Y, and Z offsets. If the accuracy is not within acceptable limits, the robot needs to adjust the camera's pose and position, realigning the camera until it is within a certain accuracy range.
The combination of PLR and PLB vision systems with robots has broad application prospects in the automotive industry, machinery manufacturing, and parts supply sectors. The in-depth application of these technologies will inject new vitality into robot applications and subsequently lead to industrial structure optimization. SICK China can leverage this market environment and its own product advantages to strengthen the promotion of this technology, thereby achieving a win-win situation with its customers.
